Crime Prediction and Reporting System

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 227
Crime Prediction and Reporting System
Mahesh Kate1, Nitin Jadhav2, Suyash Mhatre3 Dipashree Sonawale4
1,2,3 B.E. Student, Department of Information Technology, MGMCET, Kamothe
The vast expanding technology in the face of Artificial
Intelligence and Machine Learning has seen exponential
growth over the last decade. This growing change
contributes to help humankind in numerous fields
pertaining entertainment, money or finance, e-platforms
and many more. Thus, there comes a bigger picture or
issue which needs to be addressed using such
extraordinary technologies which is Crime. Overpopulated
countries like India have to withstand a lot of issues
regarding the rate at which the crimes are increasing. The
digitalization age brings us a good number of
opportunities to look into this matter and try to eradicate
it with these increasing IT advancements. The system is
proposed with a view to accommodate all the relevant
bodies which are concerned with crime in a single system.
Technically speaking, the people involved in a particular
crime solving are usually the Invigilators or the officials, a
Non- Government Organization (NGO), and a person who
has reported the crime. Thus, the system has to be planned
in such a way that an end-user like us or a normal citizen
can log-in and report a crime based on various filters used
like location and type of crime. Also, the user has to
provide the basic authentication details required for a
false/ fake complaint. Generally, there is a basic
conception that a child worker is seen working on a tea-
stall where people visit daily and they don’t do much
rather experience it and carry on with the schedule.
People find it a bit mainstream to officially lodge a
complaint for this tier of crime. The system makes it a
matter of minutes to take such hideous acts in the view of
Invigilators and ultimately helping the young lad, thus
providing free education by various NGOs associated with
our system.
Fig. 1 Prediction Graph
Also, other stakeholders like Invigilators, local authorities
and NGOs get various insights from the unstructured data
without any significant meaning. Crime visualizations
using traditional Machine Learning models provide us
with various patterns in the crime data structured
according to the need of the user. Technically, the
visualizations can be obtained according to state-wise
plots or date-wise plots. The prediction system comes in
the play when we want to emphasize the significant
problem. The system tries to predict the hotspots of crime
i.e., signaling out the regions across all India where the
next wave of surge is to be seen. Also, the crime rate of
future years can be obtained at a particular % of accuracy,
thus gaining a similar insight based on the number.
The main idea behind the project is to ease the policing
and NGO work by bringing Machine Learning and AI into
the picture by elaborating the crime patterns and rate
which could be seen in the near future and also predicting
the hotspots where the next wave is likely to occur. Thus,
easing the process of patrolling and concentrating on such
localities with more intent. The basic inspiration is that no
existing systems perform on the strategy of bringing these
stakeholders under a platform and solve the issue using
the wonderful field of Data Science. Therefore, the project
benefits the Public sector, Government sector as well as
the Private Sector.
4Professor, Mahatma Gandhi Mission's College of Engineering and Technology Kamothe, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The increasing rate in criminal activities is a
growing concern for any particular country/region. The
intention of the proposed system is to develop a web
application which is user-friendly to the stake-holders such
as invigilators, NGOs and end-users. The system establishes
a simple relation between the above-mentioned stake-
holders where any individual user can report a crime
without himself going to the police station. The invigilator’s
then can track the complaint and give an option to the NGO
for Rehab. Also this system analyzes crime data of India
scrapped through various websites. The main focus is to
predict the crime which is most likely to occur in the near
future using various Machine Learning models.
Key Words: Criminal activities, Invigilators, NGOs,
Police station, Analyze crime data
1. INTRODUCTION

The system uses the dataset provided by National Crime
Records Bureau which is the official organization designed
for crime datasets in India with a view of empowering
Indian Police with Information Technology. Thus, the data
provides the stats for various crime types from 1990 to
2018. The data is scrapped using web-scraping libraries of
Python like BeautifulSoup, etc. The data scrapped contains
redundant data, null values. Thus, data pre-processing is
implemented on the obtained dataset in order to clean the
data and replace null values using Pandas, Numpy, and
other Python Libraries. The most significant aspect of the
project is choosing a Machine Learning Algorithm that
best fits the dataset and provides maximum accuracy.
Thus, the optimum algorithm provides the prediction
which is represented using simple visualization charts and
plots. The data integrity and security is maintained
throughout the process. The system uses technical
platforms like tableau, flask, php, etc. The visualizations
and charts are shown using a couple of javascript libraries
like plot.js and chart.js. The detailing and choosing of the
most optimum machine learning algorithm is covered
briefly further in this chapter.
The major applications of the system can be providing
patterns from the large number of crime data stored under
local authorities of India. Also, Researchers can come and
find various insights required for their work in crime
related fields. Anyone registered on the platform can view
the visualization and dashboards of its choice. This can
also help in deploying a capable officer of high tier in a
region where the surge is obtained. Thus, easing or
preventing the crime even before it takes place or
reducing its strength. The Non-Government Organizations
which are associated with the system can increase their
rehab rates by focusing in such regions. Also, the crime
where Child labor and domestic violence is involved can
be taken into consideration by NGO bodies which function
for such victims. To make it even more efficient, the
system tries to provide a separate domain in Cyber
security for Domestic Violence where the basic idea is to
make people aware of the foreseen threats and prevention
measures. Also, what to do when one is caught in the trap
of cybercrime, various threat scenarios and defense
strategies protecting the passwords, secure web, etc. Such
functionalities make an attempt to contribute in the field
of cyber security. The prediction system accepts “Year”,
“Crime Type” and “State (region)” as the inputs and
outputs of the crime rate prediction up to the input year,
using state-based aggregated comparison using various
visualization techniques like chart and plots. The
visualization system consists of three main factors
namely- State Comparison for Crime against women, State
Comparison for Crime against children and State
Comparison for Crime against Indian Penal Code (IPC).
The data available is limited, thus as the data increases
further the system tries to improve its accuracy and
persistence.
2. Comparative Analysis
understand the severity of crime threat and its different
states/districts/regions of the country.
Police: The lack of structured data available to the officials
makes it difficult to detect criminal activities.
Public: No such system exists where a normal citizen can
opt for real estate in future based on the crime rate in that
particular area at that time
NGO: lack of reporting of minor cases like child labour,
domestic violence, etc makes it difficult for the NGO’s of
such type to perform effectively
Crime prediction, prevention and detection with data
mining is an exciting new area, which brings together the
disciplines of statistics, machine learning, artificial
intelligence, criminology, psychology and database
technology.
Traditional
Approach
Proposed System
Reporting
System
Reporting a case of
minor crime like
child labour,
domestic violence,
etc person can call
a helpline number
or directly report
to police station
The proposed
system allows
the user to bring
such types of
crimes in the
eyes of
respected
authorities by
just logging onto
the system.
Crime is one of the biggest and dominating problems in
our society and its prevention is an important task.
Daily there are huge numbers of crimes committed
frequently which need to be brought under control. Police
analysts are required to solve the complexities in raw data
to guide concerned authorities in arresting offenders and
implementing crime prevention strategies. However, a
huge number of crimes are being committed and the
awareness of modern criminals make it a difficult task.
The ability to analyze this amount of data with its inherent
complexities without using computational support puts a
strain on human resources.
The Proposed system scales to the scope of benefiting the
following sectors:
Government: The lack of systems or patterns to

Bringing
Crime
related
parties
together
After a Crime is
reported or
previous backlog
crimes are usually
assigned to specific
teams of officials.
All the parties
related to crime
like Invigilators,
NGOs, Citizens
are brought
together under
one system
Visualiza
tion of
Crime
data
All the crime
related data is
stored in
unstructured
format with lots
of errors and
missing values.
The unstructured
data is visualized
using plots and
graphs and stored
systematically. Thus
it can be helpful for
Invigilators, NGO’s ,
Researchers, etc
Efficiency
of NGO
The NGOs find it
difficult to get
exact
information
about the crime
and thus it is
difficult to
develop its
rehab models
Due to ease in
reporting a Crime
and readily
available data, NGOs
can efficiently
develop the Rehab
models
2. The Proposed System
The above system is proposed in such a way that it should
accommodate all the relevant bodies which are associated
with crime. The system includes access for end users like
normal citizens, Non-Government Organizations and
Invigilators of the Crime department or the official
authorities in this field. The above stakeholders of the
system need to register or sign up on the web portal in
order to access the system. The user credentials are stored
on the databases where entities for User, NGO and
Invigilators are separated. Other Databases consist of
tables that store NGO details, Invigilator details and User
details all simultaneously managed by Admin. The User
page is a friendly interface to register and report a crime
or complaint based on several filters like crime type,
region, etc. Also, users can have access to visualizations
and prediction systems where a User can check about the
crime hotspots or a likely surge in a specific region. Thus,
making him/her alert to the possibility and ultimately
creating a sense of awareness.
Fig.2 System Architecture
A researcher can also try and obtain insights from the
visualizations and predictions which can help in the work
of cyber security or crime. The user’s complaint gets
shown up on Invigilator's interface and hence the location
of the crime is popped with a marker on the map portal
which is available for Invigilators. After rescuing a victim
of a crime that is of the NGO relevant caliber, the NGO
database gets updated for a rehab mission. Thus, a child
labor victim or Domestic violence victim can be rescued by
Invigilators and now can be sent for rehab through
registered NGOs. Therefore, as the system grows more and
more Invigilators can be logged into the system to
contribute in the process of eradication of crime. Also,
large numbers of relevant NGOs can apply themselves to
the system ultimately helping the victims to not fall prey
to such malpractices again and thus carry out a successful
rehab program. The next part of the system architecture
comprises mainly two parts i.e. Predictions and
Visualizations. The major types of crime that occurred
against women are classed as a single crime named Crime
against Women. Similarly, the other three significant
classes of crimes are Crimes against Children, IPC Crimes
based on population and SLL Crimes. Any crime class
needs parameters such as “Select year”, “Select Crime
Type” and “Select State” as inputs to get predictions
according to the data as outputs. The outputs are mainly
graphs and other analytical factors. Thus, the visualization
part mainly consists of Comparisons and insights of
various crimes according to the state. The significant
factors considered for analytics are the crime rates per
one lakh people and literacy rates of Indian states. Thus,
this is practiced with such a view that the crime rates can
be depicted based on the population and also the literacy
of that region. Thus, population and literacy can be
considered the major factors which affect the crime rate of
a state. A certain value is set as a threshold for prediction
purpose as not all standard algorithms are suitable for this
predictions.

3. CONCLUSION
The outcomes and results of the above research approach
will be to design a central system with a dedicated view
and purpose i.e. Eradication of crime in India. Thus, the
system will try to throw some analytics related to crime in
the eye of the officials and related bodies. The patterns
and insights which are not seen through the raw data
stored in crime databases is made now made more clearer
and hence it ultimately helps in solving the crime at a
quicker rate. The results can be displayed using various
visualization techniques and they are showed as the
relation between two or multiple factors like Crime type
and state, etc. It has been observed that the crime rate
goes on adding numbers year by year in a populous
country like India. Thus, there is an immediate need to
take possession over this increasing malpractices by
introducing Intelligence and Machine Learning
technologies, which are the most trending fields in IT
industry right now. Hence, with the growing concern of
Cyber Crimes emerging in addition to other types of
crimes, the system also contains a dedicated guidelines to
help citizens not get trapped or fall pray. The aim was to
transform people into Cyber Secure Users. Although, there
have been numerous existing systems working in this
matter but this system stands apart due to its principle of
bringing all the stake holder relevant to crime together at
a single platform.
4. REFERENCES
[1] Tushar Sonawanev1, Shirin Shaikh2, Shaista Shaikh3,
Rahul Shinde4, Asif Sayyad5, Crime Pattern Analysis,
Visualization and Prediction using Data Mining, IJARIIE
International Journal of Advanced Research and
Innovative Ideas in Education, Vol-1 Issue-4 2015, IJARIIE-
ISSN(O)-2395-4396 Methodology for crime pattern
analysis.
[2] Dr. Zakaria Suliman Zubi1, Ayman Altaher Mahmmud2,
1Sirte University, Faculty Of Science, Computer Science
Department Sirte, P.O Box 727, Libya, 2The Libyan
Academy, Information Technology Department, Tripoli,
Libya, Using Data Mining Techniques to Analyze Crime
patterns, ISBN: 978-960-474-350-6.
[3] Jacopo Staiano,Druno Lepri, Andrey Bogomolov, Nuria
Oliver, “Once upon a crime : towards crime prediction
from demographics and mobile data”, 2014 ACM book
chapter.
[4] Neeru Mago, Design and Implementation of Intelligent
Crime Information and Analysis System (ICIAS) based on
Crime Data Mining,IJARCSSE International Journal of
Advanced Research in Computer Science and Software
Engineering, Volume 6, Issue 1, January 2016 ISSN: 2277
128X.
[5] Ruth Wario, Ivan Niyonzima, Emmanuel Ahishakiye,
Elisha opiyo omulo , “Analytics of Business Intelligence
Techniques on crime prediction”, 2017 International
journal of computer and information technology.
[6] Shiju Sathyadevan, & Surya Gangadharan.S, “Crime
analysis and prediction using Data mining”, 2014 IEEE
Journal article.
[7] Kang, H. and Kang, H, “Prediction of crime occurrence
from multi-modal data using deep learning”, 2017 ACM
Book Chapter.
[8] Ahamed Shafeeq B M1, Dr. Binu V S2, Spatial Patterns
of Crimes in India using Data Mining Techniques, Certified
International Journal of Engineering and Innovative
Technology (IJEIT) Volume 3, Issue 11, ISSN: 2277-3754
ISO 9001:2008, May 2014.
[9] Eugenio Cesario, Charlie Catlett, & Domenico Talia,
“Fore-casting Crimes Using Autoregressive Models”, 2014
IEEE Book Chapter.
[10] Vineet Pande, Viraj Samant, Sindhu Nair, “Crime
Detection using Data Mining” , 2016 International Journal
of Engineering Research & Technology (IJERT) ISSN:
2278-0181, Vol. 5 Issue 01, January-2016.
[11] Soon ae chun, venkata paturu, Rohit phatak,
Vijayalakshmi Alturi, Nebil Adam, “Crime Prediction Model
using deep neural network” , 2019 ACM Book Chapter.

Crime Prediction and Reporting System

Recommended

Recommended

More Related Content

Similar to Crime Prediction and Reporting System

Similar to Crime Prediction and Reporting System (20)

More from IRJET Journal

More from IRJET Journal (20)

Recently uploaded

Recently uploaded (20)

Crime Prediction and Reporting System